Neonate Blood Sugar Problems with Nova Statstrip meters

What did Nova say about it?

Nova thinks its user error and sent me a guide on how to collect.  The nurses that I observed are collecting according to the info sent.  Still getting complaints.

What are the complaints regarding?   

The values they are getting on the meters aren't matching the lab values.  CLSI states they have to be within 12 points of each other and they are.  Technically, the meters are working correctly.  They are complaining because their policy states that when a sugar is below a certain number, the baby needs to be admitted to NICU.  For example, the cut off for NICU is 30.  They do a POC glucose and the meter reads 25.  When they send a glucose for lab, the values are coming back 33.  

My OB, said if they have a low Neonate glucose and they repeat it, the result goes up.  I told them I would expect that.  Glucose is released when the body is stressed, fight or flight, so when you re-stick the baby the glucose has gone up due to the stress of the first stick.  OB is still concerned.  I did a cal ver on their meter, it was fine, ran QC, it was fine, I gave them another meter to appease them and they were happy.  

I explained that glucose will increase with each stick and have given them multiple meters but they still aren't happy.  

Cal and QC work fine.  I'm at a loss on what to do at this point. 

I sent them multiple research articles and ask them to do a study internally.  Collect both samples at the same time, MG and Lab, and run the MG on a final drop.  They did not do the comparison.  They accepted the research and now repeat MG on the Nova when they feel it is necessary. 

We do use RALS.  

If I'm understanding correctly, you got RALS to correct the bias on sugars under <60.  When a result comes into RALS, it corrects the bias to be more on par with the lab analyzer?  Is that correct?

We were getting a lot of <10 mg/dl so NICU was having a fit.  Yes, we ran a comparison between the morning blood gases on our ABL835's (it has whole blood glucose) and immediately tested on the Nova meter.  We ran a lot of comparisons so that we could pick out just the ones below our critical limit.  We sent the comparison data to Nova and they agreed that the negative bias on the low end was around 10 mg/dl, the slope was 1.0,  So we did put in an offset into RALS that pushed out to all the meters and haven't had complaints since.

Reine, can you tell me how you performed the comparisons?  How many samples did you use?

I'm thinking about doing a comparison myself and talking to Nova about it.  We have too many neonates being admitted to NICU unnecessarily. 

I don't know exactly how many samples we compared but we ended up with 50 comparisons with results <100.  I can send you a copy of our comparison with and without the offset.  You'll have to run your own study.  We were lucky because the main lab glucose is turned on in the ABL835s so it didn't interrupt lab testing much, they just had to take the leftover sample and test it on the Nova meter.

Yes, please.  I would love to look at your data to see how I need to approach this.

Don't even get me started on their biases on the creatinine strips.
This is super helpful information, thank you everyone that posted.
That's amazing that you can program an offset into RALS that is triggered based on initial glucose value that it will apply selectively. I don't have RALS or TELCOR and wish I did... I'm blown away by that solution.